Ricardo Lanfredi, Ambuj Arora, Trafton Drew, Joyce Schroeder
We lifted 15 functions out of this paper's own repositories and ran 10 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| ricbl/etsaliencymaps | canonical | 10 of 15 |
| Function | Status | Where it lives |
|---|---|---|
| apply_windowing | Ran | ricbl/etsaliencymaps/src/get_segmentation_baseline.py code served (permissive licence) · get_code("fc8f6600673f7e43") |
| create_heatmap | Ran | ricbl/etsaliencymaps/src/generate_heatmap_eyetracking.py code served (permissive licence) · get_code("6015f5e0532cebea") |
| find_nearest | Ran | ricbl/etsaliencymaps/src/get_segmentation_baseline.py code served (permissive licence) · get_code("1ed2739588f5bbbe") |
| get_32_size | Ran | ricbl/etsaliencymaps/src/dataset.py code served (permissive licence) · get_code("8a4f89cc27017275") |
| get_auc | Ran | ricbl/etsaliencymaps/src/dataset.py code served (permissive licence) · get_code("a0c2734a9e0a5ea7") |
| get_cases | Ran | ricbl/etsaliencymaps/src/compare_heatmaps.py code served (permissive licence) · get_code("8699ea53bda34972") |
| get_gaussian | Ran | ricbl/etsaliencymaps/src/generate_heatmap_eyetracking.py code served (permissive licence) · get_code("54967f11ccbbb5ab") |
| smooth_auc | Ran | ricbl/etsaliencymaps/src/compare_heatmaps.py code served (permissive licence) · get_code("2ba449d77622a6dd") |
| smooth_shuffled_auc | Ran | ricbl/etsaliencymaps/src/compare_heatmaps.py code served (permissive licence) · get_code("c07a0a6f9b9f548e") |
| sorter | Ran | ricbl/etsaliencymaps/src/get_center_bias.py code served (permissive licence) · get_code("c0c3f55297e9cd25") |
| crop_or_pad_to | Not yet run | ricbl/etsaliencymaps/src/generate_heatmap_model.py code served (permissive licence) · get_code("d0df7bad46f570f0") |
| getImgList | Not yet run | ricbl/etsaliencymaps/src/mimic_generate_df.py code served (permissive licence) · get_code("c49a60e9b322c63c") |
| get_filepaths | Not yet run | ricbl/etsaliencymaps/src/generate_heatmap_model.py code served (permissive licence) · get_code("df6a7d611455250b") |
| get_model | Not yet run | ricbl/etsaliencymaps/src/get_model.py code served (permissive licence) · get_code("5f0de177617b6965") |
| pre_process_path | Not yet run | ricbl/etsaliencymaps/src/dataset.py code served (permissive licence) · get_code("dd52fd91fdec8583") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
The interpretability of medical image analysis models is considered a key research field. We use a dataset of eye-tracking data from five radiologists to compare the outputs of interpretability methods and the heatmaps representing where radiologists looked. We conduct a classindependent analysis of the saliency maps generated by two methods selected from the literature: Grad-CAM and attention maps from an attention-gated model. For the comparison, we use shuffled metrics, which avoid biases from fixation locations. We achieve scores comparable to an interobserver baseline in one shuffled metric, highlighting the potential of saliency maps from Grad-CAM to mimic a radiologist's attention over an image. We also divide the dataset into subsets to evaluate in which cases similarities are higher.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2112.11716")
get_code_for_paper("2112.11716")
have("2112.11716")
Connect an agent — have() is free.